I will cover coronavirus separately. In the meantime, a lot has been happening in genomics more broadly. I’ll draw your attention to:

1. Ongoing controversy about Precision Medicine. Two pieces question the prominence it receives Precision medicine: course correction urgently needed and Promises and perils of using genetic tests to predict risk of disease. Meanwhile The Personalized Medicine Coalition has helped put together a new bipartisan caucus to advance support for personalized medicine in Washington.
2. A host of new common disease/complex trait work (separated into its own section below), including GWAS for self-reported childhood maltreatment, income, alcohol intake, lifespan. A few papers that have looked at including polygenic scores in clinical models, with mixed results.
3. A remarkable amount of legislative efforts, including a Bill on the Governor of Florida’s desk that would close some GINA loopholes, and efforts in Utah, New Jersey, and California
4. DNA from detained immigrants is being added to the federal national DNA database, CODIS, which was originally established to track violent federal offenders

 

Science

• From October, prime-editing, akin to “search and replace” rather than “cut and paste”, which doesn’t require a double strand break “Prime editing substantially expands the scope and capabilities of genome editing, and in principle could correct up to 89% of known genetic variants associated with human diseases.” In the first few weeks post the publication of this article the researchers were inundated with requests for constructs.
• Some updates from All Of Us: 339,000 consented participants, 265,000 completed the full protocol, 50% from racial and ethnic minorities, 80% from populations underserved in biomedical research. “This is more than just a data resource. It’s an ecosystem” that includes data analysis tools and participant outreach.
• Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium, reporting on the matched tumor-normal sequencing of 2,658 individuals. Key findings included: average of 4-5 driver mutations per cancer; only 5% tumors lacked any identified driver mutation; large rates of  chromoplexy (17.8% of tumours) and chromothripsis (22.3%); change of mutational signature over time in 40% of tumors; driver mutations can occur years before cancer diagnosis. 
• The NHGRI have published their draft Genomics 2020 strategic plan. They invite comments.
• Anti-CRISPRs exist in the wild, as part of the viral counter-attack to the bacterial counter-attack that is the main function of CRISPR systems in bacteria. There are also several other attempts to produce a “stop button” for gene editing, which should have as a by-product editing which can be better targeted to particular cell types and to reduce off-(sequence)-target effects. (Link to an easy to read nature news feature summary).

 

Polygenics: Common diseases/complex traits

• Two studies in JAMA showing no and limited improvement in a clinical risk score when including polygenic scores for cardiac disease
• A study in the Finnish Biobank of polygenic scores shows increased predictive power over clinical models for T2D, Breast cancer and prostate cancer, and improved reclassification over clinical thresholds for all diseases tested (those plus coronary heart disease and.atrial fibrillation)
• A meta-analysis of 42,000 individuals found “significant associations between childhood psychopathology and adult polygenic scores of major depression, subjective well-being, neuroticism, insomnia, educational attainment, and body mass index but not bipolar disorder.”
• Lifespan heritability is estimated to be 23-33%. Some of this comes from the effects of rare protein truncating variants.
• Looking across the UK, Japan, and Finland biobanks, researchers looked at how polygenic scores affected lifespan, claiming that this shows “new value of PRSs in prioritization of risk factors that could be potential targets of medical treatment to improve population health.”
• GWAS of alcohol intake.
• GWAS for income, estimating 7.4% heritability, and constructing a polygenic score capturing 1.2-2.0% of the variance in household income. The conclusion, “These results indicate that, in modern era Great Britain, genetic effects contribute towards some of the observed socioeconomic inequalities.” stresses that these numbers are non-zero, rather than that they are very small (which it equally could have done).
• A GWAS for self-reported childhood maltreatment. Why, you might ask? “A clearer understanding of the genetic relationships of childhood maltreatment, including particular abuse subtypes, with a range of phenotypes, may ultimately be useful in in developing targeted treatment and prevention strategies.” They speculate about a possible role for FOXP variants they identified in “externalizing traits”, which include overactivity, poor impulse control, noncompliance, and aggression.  
• A polygenic based score for stroke identified 0.25% of the population at threefold increased risk. This is much less discriminatory than , for example, coronary artery disease. 
• A computational method to improve polygenic score accuracy in admixed individuals, which basically a) breaks an individual’s genome into ancestry derived blocks, b) applies the ancestry relevant effect sizes to that block. This should lead to less biased estimates even if good GWAS data is only available for one of the ancestries (European).
• A preprint focused on making polygenic scores more interpretable, through identifying genetic drivers of risk: “For example, we find that BMI polygenic risk factorizes into distinct components relating to fat-free mass, fat mass, and overall health indicators like sleep duration and alcohol and water intake. Most individuals with high dPRS for BMI have strong contributions from both a fat mass component and a fat-free mass component, whereas a few ‘outlier’ individuals have strong contributions from only one of the two components.” 
• Genomic Prediction, the company that will screen embryos for polygenic scores, published a piece reporting that for Type 1 Diabetes, a 45-72% reduced risk using a polygenic score when choosing between an affected/unaffected sibling pair compared to random. Contrast this study with the article Screening human embryos for polygenic traits has limited utility which focused on expected gain, i.e. enhancement, rather than screening out high genetic risk for a disease. 
• Intuitively, if you have less accurate phenotyping, you’re less able to identify genetic effects. Proof of that for depression, where use of “minimal phenotyping” led to identifying different genetic architecture than full on medical diagnoses.
• A review about unpicking the causal effects of genetics in genotype-phenotype correlations. Population stratification, assortative mating, and dynastic effects are all captured in these correlations.
• The authors of the GWAS of same-sex behavior defend their research, arguing first that their phenotype is health related, and second that “We believe that when broad biobank consent is used, institutional bodies (such as the UK Biobank’s Board and its Access Sub-Committee) must take responsibility for approving specific research requests rather than expecting participants to understand all such details for possible research projects.” 

 

Applications

• Ancestry has laid off 100 people, 6% of its workforce. 23andMe also laid off 100 people, 14% of their workforce. Both cite declining consumer demand. Veritas halted its US operations due to financing issues.
• The Pentagon warned military personnel not to take DTC genetic tests: GINA’s protections do not extend to military personnel: “The unintentional discovery of markers that may affect readiness could affect a service member’s career.” Separately, the data’s existence could “potentially create unintended security consequences”.
• The first use of CRISPR inside the body, for a blind patient as part of a trial of an approach by company Editas. Results will take months. 
• Apple is offering free genetic testing to all its SIlicon Valley based employees.
• A public opinion survey asked about financial expectations for contributing data to genomic databanks, 11.7% no payment, 50.6% with payment, and 37.8% never. Those who wouldn’t expect payment would pay $75 for results, those who expected payment expected $95 on top of a report. 
• The American Society for Human Genetics commissioned a public survey. I thought this was the most interesting finding: “Thirteen percent report having taken a direct-to-consumer genetic test; 8 percent had a genetic test through a hospital or research center; 5 percent had received genetic counseling; and 5 percent had participated in research requiring a blood or saliva sample.”

 

Regulation etc

• He Jiankui, of Lulu and Nana fame, and two colleagues, were sentenced to jail sentences in China for “illegal medical practice”
• A call for “An international code of conduct for genomic data.” To cover: identifiability; broad consent; return of individual findings, portability and access; withdrawal; and compelled disclosure. “A first step would be to convene a meeting to determine the topics the code would touch on, the best way to consult research participants about their needs and a decision-making process that will allow the text to be finalized in a timely way.” “further regulatory uncertainty risks stalling new genomic analyses and undermining people’s faith in scientific collaboration for the public good.”
• US federal and foundation funding has been far higher for cystic fibrosis than for sickle cell anemia. The former mostly affects European ancestry individuals, the latter African ancestry individuals.
• Bioethicists suggest a duty to reinterpret genetic data
• Establishing the International Genetic Discrimination Observatory, “a unique network of researchers and other stakeholders dedicated to researching and preventing genetic discrimination worldwide. It is forward looking, with a broad purview that includes discrimination based not only on genetic information but also on other emerging forms of predictive health data. Its broad approach to genetic discrimination is framed by recognized human-rights principles including respect for autonomy, dignity, privacy, the right to science, and the right to know or not know about genetic results.” In lieu of regulation, they mention “moratoria, administrative policies, governmental decrees, mandatory ethics regulations, and agreements between organizations and patient groups”.
• Legislative efforts
  • A Florida Bill banning genetic discrimination in life insurance, lon-term care insurance, and disability insurance, passed both houses (with very few dissenting voices) and is awaiting governor signature. Insurers can only use genetic testing results if they’re part of a medical diagnosis. It will be the first state to prohibit such use.
  • Proposed California legislation to restrict genetics DTC companies use of data
  • New Jersey has proposed legislation aimed to secure greater genetic privacy, it would make DNA information the property of the person analyzed. “With this legislation, New Jersey would join 24 other states that require informed consent to disclose genetic information. The state also would become sixth after Alaska, Colorado, Florida, Georgia, and Louisiana in explicitly defining genetic information as personal property, and second only to Alaska in extending personal property rights to DNA samples.”
  • Efforts in Utah to prevent law enforcement access to consumer DNA databases.
• We each leave a metabolite trail, not protected by GINA.
• A nice summary of GINA and employment. There hasn’t been a single successful lawsuit alleging genetic discrimination by an employer, something that the Federal law GINA protects. There have been plenty of charges files with the Equal Employment Opportunity Commission, who have filed 12 cases as a result. Most of these are not due to employers getting genetic data, but family medical history data. But GINA has worked as a privacy statute. GINA covers unmanifested conditions, and the ADA covers manifested conditions that cause disability, leaving something like prediabetes in limbo. 
• DNA and law enforcement
  • DNA from detained immigrants will be entered into the FBI’s CODIS database. This database was originally designed just for violent federal offenders but has been repeatedly expanded, including to arrestees never charged. It currently contains 20 million individuals, of which 5 million are arrestees, and 40% are black. Collecting detained immigrant’s DNA will further skew the proportion of minorities represented. 
  • Local law enforcement agencies have laxer standards about whose DNA they keep. The NYPD is going to delete samples that never led to a conviction that are at least two years old.
  • Epigenetic clock tests, which can predict an individual’s age within a few months, are drawing ethical concern around their potential use on migrant minors.